[XML-SIG] XML serialization / marshalling via DTD

george willis gwillis@mail.com
Wed, 31 May 2000 15:21:42 -0400 (EDT)


I believe that such flame wars are counter productive, and have no place in
a professional environment.  They waste time and threaten to dismiss key
topics of discussion as "shortsited".

I am aware that many use "CAPITAL CRITICISM" to shout derision at one
another.  My use of capitals was to highlight my main thought in a rather
large message. As my first message is proof that capitals can be used
without derision, your response is proof that derision does not require
capitals.

Your comments that try to dismiss this issue as "shortsited" actual give
credence to my concerns.  <bold> With even more ser/deser code designed for
a specific application, which among the heaps of choices is a good starting
point for an autonomous tool to provide this service to all of the desired
consuming tools?  Which is the best code base? </bold> (see, it's just hard
to highlight without caps.)

Save your fight for the java XML products that are threatening to take the
market share.  If you took offense, you have my sincere apologies.  If you
can and will cooperate in this discussion, I welcome your help.  If not, go
with God and my blessing.

NOW, UNTO THE BUSINESS AT HAND (used to signify a new section)
------------------------------
Fredrik has made me aware that yet more code exists in even more modules. 
Thank you Fredrik.  Does anyone have experience with this code?  Does it
ser/deser an entire containment tree?  What DTD or XML Schema standards are
supported?  Does it make use of the existing SAX/DOM/Parser tools?  How does
it compare with other approaches code found in XML Objects, XML Documnent,
etc.

Has anyone else considered an architecture where ser/deser was a seperate
interface?



------Original Message------
From: "Fredrik Lundh" <effbot@telia.com>
To: "george willis" <gwillis@mail.com>, <xml-sig@python.org>,
<zope-xml@egroups.com>
Sent: May 31, 2000 6:44:40 PM GMT
Subject: Re: [XML-SIG] XML serialization /  marshalling via DTD


> WHAT WE NEED IS THIS MECHANISM, JUST AS IT IS IN JAVA XML SERIALIZATION
> CODE, TO PROVIDE AUTOMATED SER/DESER USING THE EXISTING FOUNDATIONAL WORK
OF
> SAX, DOM, AND PLUGGABLE PARSERS, SO THAT WE MAY LEVERAGE ALL THE BENEFITS
OF
> OO REUSE THROUGH THE CONSUMING TECHNOLOGIES OF RPC, PERSISTANCE, ETC.

if you had bothered to look before you started shouting at me,
you might have noticed that xmlrpclib.py and soaplib.py provide
generic marshalling code for python data structures.

but since "reuse" obviously means "inventing yet another wheel"
in your dictionary, I can only wish you good luck.

over and out /F

George Willis
gwillis@mail.com
voice: (706)206-0091
  fax: (240)337-8593 
______________________________________________
FREE Personalized Email at Mail.com
Sign up at http://www.mail.com/?sr=signup